Healthcare professionals’ perspectives on decision-making about lung cancer screening participation: a cross-sectional study in Flanders

Abstract

Background

Lung cancer remains the leading cause of cancer-related mortality worldwide, with survival largely determined by the stage at diagnosis. While low-dose computed tomography (LDCT) screening reduces mortality, no population-based screening programme is currently implemented in Belgium, altough interest is increasing. However, its success will not only depend on participation but also on informed, well-considered choices, which are linked to better adherence and ethically sound practice. Shared decision-making (SDM) can play a central role in fostering such value-congruent participation, yet its implementation relies heavily on healthcare professionals (HCPs). Understanding how HCPs perceive their role, preparedness, and interprofessional responsibilities is crucial for developing ethically sound and feasible decision-making strategies for lung cancer screening.

Methods

A cross-sectional online survey was conducted among HCPs in Flanders involved in lung cancer prevention, detection, and care (N = 415). The questionnaire assessed perceptions of patient autonomy, professional guidance, and roles in supporting decision-making about potential population-based lung cancer screening. Data were analysed using descriptive and inferential statistics, with group comparisons across gender, profession, and experience.

Results

Participants represented general practitioners (49%), pharmacists (29%), tobacco treatment specialists (11%), and medical specialists (11%). Confidence in patients’ ability to make an informed participation decision was moderate (M = 6.08/10), while willingness to support patients was high (M = 7.93/10). More than 85% of participants preferred models in which the patient makes the final decision, either independently or after considering the HCP’s advice, suggesting broad support for patient autonomy alongside some reservations about patients’ ability to make an independent choice. Nevertheless, intentions to apply SDM-consistent decision support behaviours in response to invitees’ questions were generally high across professional groups. Familiarity with SDM, however, was moderate overall (M = 6.23/10) but varied significantly across professions and experience levels, with pharmacists and tobacco treatment specialists reporting lower knowledge.

Conclusions

HCPs report only moderate confidence in patients’ decisional capability to make an autonomous decision about participation in lung cancer screening, yet they endorse patient involvement and express strong willingness to provide decision support. Strengthening decision-support tools, training, and interprofessional collaboration structures among HCPs will be vital to translate patient-centred decision-making into routine cancer screening practice.

Clinical trial number

Not applicable.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12911-026-03441-0.

Background

Despite major advances in oncology and preventive care, lung cancer continues to account for the largest share of cancer-related deaths worldwide [1]. Survival outcomes in lung cancer are highly dependent on the stage at which the disease is detected, with substantially better prognoses when identified early [2, 3]. Yet, the majority of cases continue to be diagnosed at an advanced stage, highlighting the ongoing need for more effective early detection approaches [3, 4]. Among available methods, low-dose computed tomography (LDCT) has proven particularly effective in reducing mortality among individuals at elevated risk [4–6]. This effectiveness is supported by landmark randomized controlled trials, including NLST, NELSON, and MILD, which demonstrate that LDCT screening enables earlier detection of lung cancer and significantly reduces disease-specific mortality [4, 7, 8]. In response, numerous countries have established organised screening programmes targeting high-risk individuals, typically defined by age and smoking history [9, 10]. Belgium, where the organisation of cancer screening falls under the responsibility of the regional governments, has not yet introduced a population-based lung cancer screening initiative [11]. However, in Flanders, the ZORALCS pilot study, launched in 2025, is assessing the feasibility of implementing a population-based lung cancer screening programme [12]. Complementary research by García et al. has shown that potential participants in Flanders are generally receptive to screening, particularly when the programme is reimbursed and accompanied by targeted educational initiatives for individuals with lower socioeconomic backgrounds [13].

While public receptiveness is encouraging, willingness alone does not guarantee meaningful or sustained participation. For lung cancer screening to achieve its intended benefits, individuals must be supported in making informed, value-congruent choices and in adhering to recommended follow-up [14]. Shared decision-making (SDM) can play a central role in this process, as it can enhance patients’ understanding of both the potential benefits and harms of low-dose CT screening, enabling decisions that align with personal values and fostering sustained adherence [14–16]. Evidence indicates that SDM improves decision quality and adherence to annual follow-up, the latter being essential for achieving the mortality benefits associated with screening [17, 18]. Recently, Elwyn et al. conceptualised SDM as a collaborative and iterative process in which healthcare professionals (HCPs) invite individuals to participate, support understanding of available options and their consequences, and allow time for deliberation that respects patients’ values and preferred level of involvement [19]. Importantly, SDM does not require consensus or equal decisional responsibility, but rather aims to provide appropriate support while acknowledging power asymmetries and patient vulnerability [19]. This broader conceptualisation is particularly relevant in population-based screening contexts, where participation decisions may occur outside traditional clinical encounters and HCPs may be consulted on demand rather than acting as gatekeepers. However, implementing SDM in routine clinical practice will demand considerable commitment and skill from HCPs, who are central to facilitating patient involvement in this process. To implement SDM effectively in the context of lung cancer screening, it is essential to understand how HCPs themselves perceive and approach it. As key enablers of SDM, their active engagement is crucial to move the concept from theory into practice. Yet, the realities of healthcare delivery, including limited time, competing priorities, and resource constraints, may hinder the practical implementation of SDM in daily clinical practice [20, 21].

In a potential population-based lung cancer screening programme in Belgium, where population-based screening is organised regionally, several core steps, such as invitation, risk assessment, and eligibility verification, are likely to be organised centrally, in line with existing screening programmes. As a result, HCPs would not function as gatekeepers for screening participation. Instead, they may be consulted when invitees seek clarification or support in weighing the potential benefits and harms of participation. In this context, participation decisions are expected to occur largely outside opportunistic clinical encounters. SDM is therefore conceptualised as an on-demand process in this setting, applied when individuals seek guidance rather than as a discipline-specific clinical act. Any consulted HCP may apply SDM-consistent communication and deliberation principles to support informed and value-congruent participation decisions within the boundaries of their usual professional practice.

Identifying how HCPs perceive their role in helping patients decide about participation in screening, and to what extent they are prepared to assume this responsibility, is therefore crucial. Understanding these perspectives will help define a realistic and sustainable approach, a “middle ground” that respects both the ethical principles of informed choice and the practical constraints of healthcare delivery. Accordingly, this study explores how HCPs in Flanders, a region of Belgium, conceptualise the decision-making process surrounding individuals’ participation in lung cancer screening, providing essential insights to inform the design of ethically sound and practically feasible screening policies.

Methods

This quantitative, cross-sectional survey study aimed to explore how HCPs in Flanders perceive and approach decision-making in the context of a potential population-based lung cancer screening programme. Specifically, the study examined how HCPs perceive patients’ decision-making autonomy, their own role in supporting informed and value-congruent choices, and interprofessional differences in these perceptions. Ethical approval for the study was obtained from the Ethics Committee Research of University Hospitals Leuven (UZ Leuven)/ KU Leuven (S69748).

Survey development

A self-administered survey was developed for HCPs involved in the prevention, early detection, and treatment of lung cancer. Its structure and content were guided by insights from existing literature, which helped identify core themes related to clinical decision-making and professional involvement in the context of a potential population-based lung cancer screening programme. In this study, ‘decision-making’ refers specifically to the participation decision (whether to participate), not to subsequent clinical decisions related to diagnostic work-up, follow-up imaging, or invasive procedures.

Participants were first provided with an introductory information sheet describing the study. The questionnaire itself consisted of two sections: (i) demographic and professional background questions, and (ii) items related to clinical decision-making in the context of lung cancer screening. In addition, a third section addressed HCPs’ broader perspectives and attitudes toward the implementation of a population-based screening programme; these findings are reported separately in a complementary publication.

The survey mainly consisted of closed-ended, multiple-choice questions, with some allowing for multiple answers. It also incorporated several Likert-scale items to assess respondents’ level of agreement with specific statements. In addition, two validated scales from the literature were incorporated to ensure conceptual robustness [22–24]. The web-based survey was developed using the LimeSurvey platform and designed to be accessible across multiple devices. Question order and grouping were carefully structured to reduce cognitive load and ensure a logical flow through clear thematic organisation and smooth transitions.

A formal pilot test was not undertaken due to the brevity of the survey and the incorporation of previously validated items. Nevertheless, the complete questionnaire was carefully reviewed by the research team to ensure clarity and relevance.

Standardized questionnaires

Two validated instruments were integrated into the survey.

Shared decision-making questionnaire–physician version (SDM-Q-DOC) [24]

The Shared Decision-Making Questionnaire–Physician version (SDM-Q-Doc) is a 9-item instrument that assesses HCPs’ perceived extent of SDM during clinical encounters [24]. In the present study, the Dutch validated version was used with wording adapted to reflect a hypothetical lung cancer screening context, given that population-based screening has not yet been implemented in Flanders [22].

Respondents were asked to indicate how they would ideally apply these behaviours when an invited person seeks support regarding participation in lung cancer screening. Accordingly, the adapted SDM-Q-Doc was intended to capture respondents’ intentions to apply SDM-consistent decision support behaviours within their professional scope (e.g., supporting understanding, discussing options, and facilitating value clarification) in response to participation-related questions, rather than responsibility for eligibility assessment, referral, or programme entry.

Control preference scale (CPS) [23]

A modified version of the CPS was included to evaluate HCPs’ preferences for decision-making roles in the context of lung cancer screening. Originally designed to assess patient preferences, the scale was adapted to reflect HCPs’ views on ideal decision control distribution. The response options ranged from fully patient-led to fully physician-led decisions, with SDM as the midpoint [23].

Recruitment and participants

To be eligible for inclusion, participants had to be currently practising in Flanders, professionally active in the prevention, early detection, or management of lung cancer, and at least 18 years old at the time of recruitment. The survey targeted four main professional groups: general practitioners, pharmacists, tobacco treatment specialists, and medical specialists engaged in lung cancer–related care. Recruitment followed a purposive and snowball sampling strategy, using digital flyers, email invitations, and social media outreach distributed via professional networks, sector-specific journals, and relevant stakeholder organisations. Data collection was conducted between November 2024 and April 2025. Participation was entirely voluntary, and no financial incentives were offered.

Survey data analysis

All survey data were analysed in IBM SPSS Statistics, version 29.0, using a combination of descriptive and inferential statistical techniques. Participants were included in the analytic dataset if they completed at least one item beyond the demographic section. Respondents who provided only demographic information were excluded. Because the survey incorporated conditional branching and some participants discontinued partway through, the number of valid responses differed across items. Consequently, statistics are presented on an item-by-item basis, according to the available response count.

Inferential tests were selected based on the nature of the variables, data distribution, and sample size to ensure appropriate statistical analysis. Depending on these characteristics, non-parametric or parametric tests were employed (see Additional File 1, Supplementary Table 1). All tests were two-tailed. To maintain analytical consistency and prevent selective subgroup analysis, comparisons were systematically performed across gender, professional role, and years of work experience, independent of prior expectations or apparent trends. Although detailed information was collected on individual medical specialties, limited numbers in some subgroups restricted separate analyses. To preserve statistical robustness, HCPs were therefore classified into four broader professional categories: general practitioners, pharmacists, tobacco treatment specialists, and medical specialists, with the latter grouping multiple specialties. Similarly, professional experience was divided into four categories, junior (1–5 years), mid-career (6–14 years), expert (15–24 years), and senior (≥ 25 years), based on the distribution of responses.

To control for potential Type I error inflation due to multiple testing, Bonferroni-adjusted significance thresholds were applied per item, proportional to the number of pairwise comparisons conducted within each grouping factor. A nominal alpha of 0.05 was used for overall comparisons, with Bonferroni-corrected alpha levels applied to post hoc pairwise analyses. Only results meeting the corrected significance threshold (p < adjusted α) are reported in the Results section; full outputs, including non-significant results, are provided in Additional File 1.

In addition, a single exploratory correlation analysis was conducted to examine the relationship between HCPs’ perceived patient capability to make an independent decision about screening participation and their willingness to support that decision. This analysis was selected based on theoretical relevance and the significant attitudinal patterns observed in prior group comparisons. Because both variables were measured on bounded continuous scales and did not consistently meet normality assumptions, Spearman’s rank-order correlation was applied. Given the exploratory nature of this analysis, no Bonferroni correction was implemented.

Results

Participant characteristics

Survey data from 415 HCPs were included in the final analysis. The sample comprised 305 females (73%) and 110 males (27%), with a mean age of 41.1 years (SD = 12.5; range = 24–75). Participants represented a broad spectrum of professional backgrounds, including general practitioners (49%), pharmacists (29%), tobacco treatment specialists (11%), and medical specialists (11%). On average, respondents reported 16.1 years of professional experience (SD = 12.1; median = 12; range = 1–50). Of the total sample, 370 participants (89%) completed the survey in full. A detailed overview of participant demographics and professional characteristics is provided in Table 1.

Table 1.

Characteristics of participating HCPs

Characteristic	n	%
Sex
Females	305	73%
Males	110	27%
Age (years)
≤ 29 year.	92	22%
30 to 39 year.	132	32%
40 to 49 year.	73	18%
50 to 59 year.	81	20%
≥ 60 year.	37	8%
Profession
General practitioners	203	49%
Pharmacists	121	29%
Tobacco treatment specialists	44	11%
Medical specialists	47	11%
Experience (Years experience)
Junior (1–5 years)	102	25%
Middle (6–14 years)	124	30%
Expert (15–24 years)	69	17%
Senior (≥ 25 years)	120	29%

Percentages might not add to 100% because of rounding

Perspectives on patient autonomy and decision support

To gain deeper insight into how HCPs view their own and patients’ roles in making the decision whether or not to participate in lung cancer screening, the survey included several items assessing perceived patient capability, willingness to provide support, and preferred roles in the decision-making process.

The first question examined HCPs’ views on patients’ ability to make an independent and informed decision regarding participation. Respondents (n = 371) were first asked to rate, on a 10-point scale (1 = Not capable at all, 10 = Very capable), the extent to which they believed potential screening participants can make an autonomous and informed decision regarding participation. The mean score was 6.08 (SD = 1.57), with a median and mode of 6, indicating moderate overall confidence in patients’ decision-making abilities. The distribution was relatively symmetrical (skewness = − 0.13), with the majority of respondents (65.3%) assigning scores of 6 or higher. This pattern is also reflected in Fig. 1, which shows a clear peak at 6 and 7 on the scale. No statistically significant differences were found between male and female respondents (p =.431), as determined by an independent t-test, or between professional groups (p =.396), based on a one-way ANOVA. However, years of professional experience significantly influenced perceptions (F(3,367) = 4.960, p =.002). Senior professionals reported the highest mean score (M = 6.52, SD = 1.57), followed by experts (M = 6.19, SD = 1.43) and junior professionals (M = 5.84, SD = 1.55). The lowest average score was observed in the middle experience group (M = 5.80, SD = 1.59). Post hoc comparisons using Bonferroni correction (α = 0.0083) indicated that senior professionals rated patient capability to make an independent decision significantly higher than those in the middle experience group (p =.004). Although statistically significant, differences by experience were small, so interpretation should prioritise the overall distribution rather than subgroup contrasts. Overall, scores clustered above the midpoint, suggesting cautious optimism about patients’ capability rather than widespread scepticism. Full test statistics are presented in Additional file 1, Supplementary Table 2.

Fig. 1.

Distribution of HCPs ratings on patient capability to make an autonomous screening decision. Visual representation of HCPs’ ratings of how capable they believe potential lung cancer screening participants are in making an autonomous decision, measured on a 10-point scale. The height of each bar reflects the proportion of respondents endorsing each score, and colour intensity indicates relative frequency

Given this moderate confidence in patients’ ability to make an independent and informed choice, the next item explored the extent to which HCPs (n = 371) were willing to assist patients in the decision-making process. Responses, rated on a 10-point scale, (1 = Not willing at all, 10 = Very willing) indicated a relatively high overall willingness (M = 7.93, SD = 1.75), with a median of 8.00 and a negatively skewed distribution (–1.041), reflecting a general inclination toward patient support. Nearly 69% of respondents rated their willingness at 8 or higher, and only 4% provided a score of 4 or lower. Statistical analysis revealed no significant gender differences (p =.057), and no effect of professional experience (p =.206). However, significant differences were observed across professional groups (H(3) = 27.42, p <.001). Among all professions, general practitioners reported the highest mean willingness (M = 8.34, SD = 1.58), followed by medical specialists (M = 7.85, SD = 2.01), tobacco specialists (M = 7.70, SD = 1.56), and pharmacists (M = 7.36, SD = 1.83). Post hoc tests showed that general practitioners were significantly more willing to support patients than pharmacists (p <.001) and tobacco specialists (p =.004). Although statistically significant, differences across professional groups were modest, with high willingness reported across all groups. These distributions are shown in Fig. 2. No other between-group differences reached statistical significance. Full test statistics are presented in Additional file 1, Supplementary Table 3.

Fig. 2.

Willingness of HCPs to support patients in decision-making about lung cancer screening, by professional group. Bar segments indicate the percentage of respondents selecting each score on a 10-point Likert scale (1 = not willing at all, 10 = very willing). Asterisks indicate statistically significant post hoc differences: * = p <.001 (general practitioners vs. pharmacists); ** = p =.004 (general practitioners vs. tobacco specialists). Based on Kruskal–Wallis and Bonferroni-adjusted Mann–Whitney U tests

To complement these findings, the CPS was used to explore how HCPs think decision-making about participation in lung cancer screening should ideally be shared between patients and clinicians. The CPS is a five-point measure ranging from 1 (“provider-led decision-making”) to 5 (“patient-led decision-making”) that captures the extent to which respondents prefer either a patient-led, shared, or provider-led approach to decision-making. Among the respondents (n = 366), the majority (n = 187, 51.1%) preferred that the patient makes the final decision while carefully considering the provider’s opinion. Another 33.9% (n = 124) supported a model in which the patient decides independently, and only 13.7% (n = 50) favoured an SDM model. A small number preferred a more provider-led approach: 0.5% (n = 2) indicated that the provider should decide after considering the patient’s opinion, and only 0.8% (n = 3) supported full provider control. Responses clustered around categories 4 and 5, and the distribution was left-skewed (skewness = − 0.85). Higher CPS scores indicate a stronger preference for patient autonomy in decision-making. The mean CPS score was 4.17 (SD = 0.74), reflecting an overall tendency towards patient-led decision-making. No significant differences in preferred decision-making roles were observed between male and female HCPs, as indicated by a Mann–Whitney U test (p =.900) or across professional groups, as shown by a Kruskal–Wallis test (p =.460). However, a Kruskal–Wallis test suggested an overall effect of professional experience on CPS scores (H(3) = 8.416, p =.038). Mean scores were highest among middle-career professionals (M = 4.30, SD = 0.65), followed by junior (M = 4.18, SD = 0.81) and senior professionals (M = 4.12, SD = 0.77), while expert professionals scored lowest (M = 3.98, SD = 0.71). Post hoc pairwise comparisons using Bonferroni-adjusted Mann–Whitney U tests (α = 0.0083) indicated that only the difference between middle-career and expert professionals remained statistically significant (p =.004). The experience effect was statistically significant but very small, and overall patterns consistently indicated a strong preference for patient-led decision-making. Full test statistics are presented in Additional file 1, Supplementary Table 4.

Shared decision-making: knowledge, attitudes, and support

Building on these attitudinal findings, the next section focused on HCPs’ familiarity with the concept of SDM and their self-reported intentions to apply SDM principles in the context of lung cancer screening.

Participants rated their familiarity with the concept of SDM on a scale from 1 (“no knowledge at all”) to 10 (“very extensive knowledge”). The average score across all respondents (n = 370) was 6.23 (SD = 3.02), with a median of 7 and a mode of 8. General practitioners reported the highest overall familiarity (M = 7.90, SD = 2.13), followed by medical specialists (M = 6.62, SD = 2.65). Pharmacists scored the lowest on average (M = 3.70, SD = 2.56), with tobacco specialists also reporting relatively limited familiarity (M = 5.09, SD = 2.82). These differences were supported by a Kruskal–Wallis test, which indicated a significant effect of profession on familiarity scores (H(3) = 135.74, p <.001). Post hoc pairwise comparisons (Bonferroni-adjusted α = 0.0083) revealed that general practitioners reported significantly higher familiarity than medical specialists (p =.002), tobacco specialists (p <.001), and pharmacists (p <.001). Additionally, pharmacists scored significantly lower than general practitioners (p <.001) medical specialists (p <.001) and tobacco specialists (p =.006). In terms of professional experience, junior professionals reported the highest average familiarity with SDM (M = 7.50, SD = 2.52), while experts (M = 5.12, SD = 3.00) and senior professionals (M = 5.17, SD = 3.01) reported the lowest. Middle-career professionals fell in between (M = 6.76, SD = 2.89). A Kruskal–Wallis test confirmed significant differences across these experience groups (H(3) = 42.85, p <.001). Post hoc comparisons (Bonferroni-adjusted α = 0.0083) indicated that both junior and middle-career professionals reported significantly higher familiarity than experts and seniors (all p <.001). Finally, a statistically significant difference was observed between genders, with male participants reporting higher familiarity (M = 6.85, SD = 2.65) than female participants (M = 6.01, SD = 3.11; p =.037). Overall, differences in SDM familiarity were most pronounced across professional groups, more modest across experience levels, and minimal across gender, indicating that professional role is the dominant source of variation in conceptual SDM knowledge. Full test statistics are presented in Additional file 1, Supplementary Table 5.

While familiarity reflects conceptual awareness, intentions to engage in SDM reveal the readiness of HCPs to translate these principles into practice. As lung cancer screening has not yet been introduced in Flanders, measuring such intentions offers insights into potential implementation readiness. Accordingly, a modified version of the SDM-Q-Doc was administered. The original nine items were adapted to reflect a hypothetical lung cancer screening programme that has not yet been implemented in Flanders. Responses (n = 370) were scored on a 0–5 Likert scale (0 = Completely disagree, 5 = Completely agree). An overall SDM-Q-Doc total score (0–45) was calculated by summing all nine items, consistent with common practice in prior SDM-Q-Doc research [25, 26]. Higher total scores indicate a stronger intention to engage in SDM. These scores should be interpreted as behavioural intentions to provide SDM-consistent decision support within professional scope in response to invitee questions.

The mean total score across the nine items was 31.60 (SD = 6.86), with a median of 32 and a mode of 36, indicating that responses were skewed towards the higher end of the scale. Mean scores and standard deviations for each of the nine SDM-Q-Doc items are visually presented in Fig. 3 to illustrate item-level responses. Detailed descriptive statistics for all items are provided in Additional file 1, Supplementary Table 6. Internal consistency of the instrument was good (Cronbach’s α = 0.836), supporting the reliability of the scale.

Fig. 3.

Mean scores per item on the adapted SDM-Q-Doc. Mean item scores are shown for the nine adapted items of the SDM-Q-Doc. Error bars represent standard deviations. Full test statistics are shown in Additional file 1, Supplementary Table 6. The items were: (SDM1) “I would make it clear to the patient that a decision needs to be made about participation in the screening”; (SDM2) “I would want to know how the patient wishes to be involved in making the decision about the screening”; (SDM3) “I would inform the patient about the option of lung cancer screening, but also about the option of not participating”; (SDM4) “I would clearly explain the pros and cons of lung cancer screening as well as of not participating”; (SDM5) “I would help the patient understand all the information”; (SDM6) “I would ask the patient which option he or she prefers”; (SDM7) “I would thoroughly weigh the different options together with the patient”; (SDM8) “I would make a choice together with the patient for one of the options”; and (SDM9) “I would make an appointment with the patient about the next steps”

No significant differences in SDM-Q-Doc total scores were observed between male and female respondents, as indicated by an independent samples t-test (p =.206), or between levels of professional experience, based on a one-way ANOVA (p =.114). However, a one-way ANOVA revealed significant variation across professional groups (F(3, 366) = 4.72, p =.003). Post hoc Bonferroni comparisons showed that general practitioners reported significantly higher total SDM scores than pharmacists (p =.002), where other comparisons were non-significant. The overall difference between professional groups was statistically significant but small, with broadly comparable scores across groups. Mean item scores are presented in Fig. 3, and total scores by profession are shown in Fig. 4. Full test statistics are shown in Additional file 1, Supplementary Table 6.

Fig. 4.

Mean SDM-Q-Doc total scores by professional group. Bars represent mean total scores (range: 0–45) on the SDM-Q-Doc scale across four professional groups. Error bars reflect standard deviations. * indicates a statistically significant difference between general practitioners and pharmacists (Bonferroni-corrected p =.002). Higher scores indicate greater intention to engage in SDM

Exploratory correlations between attitudes and behavioural intentions

An exploratory correlation analysis examined the link between perceived patient capability to make an independent decision (1 = not at all capable, 10 = very capable) and the HCP’s willingness to support that decision (1 = not at all willing, 10 = very willing). A statistically significant but small positive association was found (ρ = 0.197, p <.001, N = 369), as visualised in Fig. 5. This finding is consistent with the idea that HCPs who express more confidence in patients’ autonomy also report slightly greater willingness to assist with the decision. Full test statistics are shown in Additional File 1, Supplementary Table 7.

Fig. 5.

Bubble chart visualizing the correlation between perceived capability and willingness to support decision

Discussion

This study provides important insights into how HCPs in Flanders perceive decision-making in the context of a potential population-based lung cancer screening programme. Specifically, it explores their views on patient autonomy, perceived decision-making capability, and the role of clinicians in supporting informed choices.

The findings indicate that HCPs expressed moderate confidence in patients’ ability to make autonomous and informed participation decisions, with an average rating just above six out of ten. This cautious confidence may reflect perceived challenges related to health literacy and the complexity of weighing potential benefits and harms, particularly in lung cancer screening, which often targets populations with greater socio-economic or educational vulnerabilities; however, such interpretations remain speculative [27, 28].

Interestingly, this moderate level of trust in patients’ decision-making capabilities stands in contrast to HCPs’ clear normative preference for patient autonomy. As reflected in the CPS scores, over 85% of respondents favoured decision-making models in which the patient either decides independently or makes the final decision after considering the provider’s advice. Only a small minority preferred provider-led decision-making. This discrepancy highlights a key tension: HCPs value patient autonomy as an ethical ideal, yet question whether patients can genuinely act autonomously in complex screening decisions, where participation involves weighing potential benefits against possible harms, and decisions should ultimately reflect individual values and understanding [17, 29, 30].

Encouragingly, most respondents expressed a strong willingness to support patients in this decision-making process. While general practitioners reported the highest willingness to engage in the decision-making process, other HCPs, including pharmacists, tobacco specialists, and medical specialists, also demonstrated substantial readiness to assist patients. This strong commitment among general practitioners is noteworthy given that time constraints and competing priorities are among the most frequently cited barriers to SDM in primary care [16]. Nonetheless, this finding may suggest that, despite these practical challenges, general practitioners view SDM as an essential part of their professional duty and are motivated to integrate it whenever feasible. Together, these results suggest that HCPs are motivated to guide patients and view decision support as part of their professional responsibility. However, the way this support is currently understood and conceptualised appears somewhat limited. Only a minority of respondents explicitly identified SDM as the ideal model in the CPS, yet their responses on the SDM-Q-Doc suggest that they value and intend to apply many of its underlying principles within a potential screening context. When presented with specific behaviours, such as clarifying options, discussing benefits and harms, and exploring patient preferences, many HCPs indicated a willingness to perform these actions if a lung cancer screening programme were to be introduced.

The discrepancy between the CPS and SDM-Q-Doc results suggests that while HCPs may not explicitly associate decision support with SDM, many nonetheless endorse key elements of the approach when reflecting on a hypothetical screening context. This discrepancy can be seen as fitting within the organisational context of a future, centrally organised screening pathway, where SDM-consistent decision support for participation decisions may be provided by different HCPs depending on where or when invitees seek guidance (e.g., general practice, community pharmacy, or smoking cessation services). Such distributed decision support reflects the on-demand nature of SDM in population-based screening in Flanders, in which key programme components (e.g., invitation and eligibility assessment) are organised centrally rather than through individual clinical encounters. Strengthening conceptual understanding of SDM and its practical application within a population-based screening pathway could help translate these intuitive practices into more structured, evidence-based decision support once lung cancer screening is introduced. Given that not all HCPs are equally positioned, trained, or resourced to engage in full SDM as traditionally conceptualised, decision coaching may represent a complementary approach. Delivered as non-directive, skills-based support, decision coaching can help individuals recognise decisional needs, understand options and consequences, clarify values, and prepare for deliberation. A realist review by Zhao et al. demonstrated that decision coaching is most effective when supported by appropriate training and organisational structures, and when mechanisms such as trust-building, tailored support, and facilitation of deliberation are activated [31]. Positioned in this way, decision coaching does not replace SDM but can support its principles in settings where HCPs contribute to informed participation decisions without assuming responsibility for clinical management or eligibility assessment.

Bridging this gap between positive intentions and more explicit conceptual understanding will be essential for effective implementation. Our findings highlight opportunities for targeted interprofessional training, clearer guidance on role division, and the development of practical decision support tools that align with SDM principles [20, 21]. Importantly, these implications follow from HCPs’ reported perspectives, role preferences, and intended decision-support behaviours in a hypothetical screening context, rather than from observed clinical encounters. Rather than assigning fixed ‘leader’ and ‘supporter’ roles, future implementation efforts should focus on enabling multiple HCPs to provide high-quality, SDM-consistent decision support on demand. Clear pathways within the screening programme should support timely escalation of medically complex questions, while preserving flexibility in where and by whom decision support is delivered.

The observed differences in SDM familiarity, with lower levels among pharmacists and tobacco treatment specialists and greater familiarity among junior professionals than among senior experts, may reflect differences in exposure to SDM concepts across training cohorts. The limited SDM familiarity among pharmacists is particularly striking, considering their accessibility and close contact with patients in community settings. This points to an important opportunity to strengthen SDM-related competencies in pharmacy education, with a focus on SDM-consistent decision support rather than on leading full clinical decision-making encounters, enabling pharmacists to play a more active role in patient education and value clarification. Community-based professionals such as pharmacists and tobacco treatment specialists could play a complementary role in patient education and decision support within their existing scopes of practice, helping to distribute the workload more evenly across the healthcare system. In addition, engaging community health workers could further enhance outreach and accessibility, since their involvement has been cited in other screening studies as a facilitator for improving awareness, supporting recruitment, and helping participants understand information and make informed decisions [20, 21]. Strengthening both knowledge and collaboration across professional groups will be key to embedding informed, patient-centred decision-making within future lung cancer screening programmes.

Strengths and limitations

To our knowledge, this is the first study in Belgium to examine the aspects of clinical decision-making in the context of lung cancer screening. By capturing the views of multiple professional groups, including general practitioners, pharmacists, tobacco treatment specialists, and medical specialists, this study provides a comprehensive and interprofessional perspective on how decision-making roles are perceived across the healthcare system. The inclusion of validated instruments such as the CPS and the adapted SDM-Q-Doc further strengthens the conceptual robustness and allows comparison with international research.

Nevertheless, several limitations should be considered when interpreting the findings. First, the use of self-report measures may have introduced social desirability bias, particularly regarding SDM-consistent behaviours. Second, voluntary participation and recruitment through professional networks may have resulted in self-selection, which could limit generalisability. The sample included a high proportion of women, which may reflect gender distributions in some healthcare professions as well as the recruitment channels used. Third, perceived patient decisional capability was assessed using a single global item referring to “most patients”, which may obscure substantial heterogeneity by age, health literacy, education, comorbidity, and psychosocial context. In addition, patient decisional capability was assessed indirectly via HCPs’ perceptions, which may not fully capture patients’ own preferences or the heterogeneity that exists across patient subgroups. Fourth, SDM-related responsibilities may not map equally onto all professional scopes of practice; results should therefore be interpreted as intentions to provide SDM-consistent decision support within a centrally organised pathway rather than as evidence of identical clinical responsibilities across professional groups. Finally, the cross-sectional design and absence of qualitative explanations limit causal interpretation of observed associations and do not allow deeper insight into underlying motivations or contextual constraints. Despite these limitations, this study offers an important first step in understanding how HCPs in Flanders conceptualise and approach decision making for lung cancer screening. The interprofessional design and focus on both attitudes and behavioural intentions provide valuable guidance for implementation planning, training initiatives, and the development of decision support strategies.

Future perspectives

Future research should build on these findings by examining more deeply how HCPs perceive and operationalise decision-making processes within preventive care. Qualitative approaches could provide valuable insight into the contextual factors that shape how HCPs communicate risks, support patient autonomy, and manage the practical constraints of implementing decision-making in routine care. Such understanding is essential to design feasible and context-specific strategies that strengthen informed and value-based decision making in lung cancer screening. Importantly, our findings indicate that SDM-consistent decision support may be distributed across professional groups within their scope of practice, while medically complex screening questions may require clinician-specific follow-up within the screening pathway. Other HCPs, including pharmacists, tobacco treatment specialists, and medical specialists, also expressed readiness to assist patients in making informed participation choices. Future work should therefore explore how this interprofessional involvement can be organised in practice, identifying effective models for collaboration, training, and continuity of patient guidance across the healthcare system. By clarifying these interprofessional roles and addressing gaps in knowledge and practical skills, future research can help translate these positive professional attitudes into consistent, patient-centred decision-making practices that enhance both the ethical and practical foundations of (lung) cancer screening.

Conclusion

This study provides novel insights into how HCPs in Flanders conceptualise decision-making in the context of a potential population-based lung cancer screening programme. Although willingness to support patients in this process was high, confidence in patients’ ability to make informed participation choices remained moderate, highlighting the need for structured decision support and practical guidance. Strengthening HCPs’ knowledge and confidence in guiding patients through informed and value-based decision-making, including the principles of SDM, will be essential to ensure patient-centred screening practice. Promoting interprofessional collaboration and developing accessible tools for decision support may help translate these principles into consistent clinical practice, supporting the sustainable implementation of lung cancer screening.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

AMSTAR 2

A Measurement Tool to Assess Systematic Reviews 2

CPS

Control Preferences Scale

HCP

Healthcare Professional

HCPs

Healthcare Professionals

IQR

Interquartile range

LDCT

Low-Dose Computed Tomography

MILD

Multicentric Italian Lung Detection study

NLST

National Lung Screening Trial

NELSON

NEderlands Leuvens Longkanker Screening Onderzoek

SDM

Shared Decision-Making

SDM-Q-Doc

Shared Decision-Making Questionnaire - Physician version

Author contributions

Conceptualisation, L.M., G.V.H., C.D., and I.H.; methodology, L.M., G.V.H., C.D., and I.H.; validation, G.V.H., C.D., and I.H.; formal analysis, L.M., and C.V.; investigation, L.M., and C.V.; resources, L.M.; data curation, L.M., writing—original draft preparation, L.M; writing—review and editing, L.M., G.V.H., C.D., and I.H.; visualisation, L.M.; supervision, G.V.H., C.D., and I.H.; project administration, I.H.; funding acquisition, G.V.H., and I.H. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by external funding from Kom op Tegen Kanker, a Belgian non-profit cancer foundation.

Data availability

All data generated or analysed during this study are included in this published article and its supplementary information files.

Declarations

Ethics approval and consent to participate

Ethical approval for this study was obtained from the Ethics Committee Research of University Hospitals Leuven (UZ Leuven) and KU Leuven, Belgium (S69748). The Ethics Committee Research of University Hospitals Leuven (UZ Leuven) and KU Leuven confirmed that the study did not fall under the scope of the Belgian Law of 7 May 2004 on experiments on the human person, as it involved an anonymous, non-interventional survey among healthcare professionals. As a result and given the anonymous and non-interventional nature of the survey, the requirement for informed consent was formally waived by the Ethics Committee Research of University Hospitals Leuven (UZ Leuven) and KU Leuven. The study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.